The present invention generally relates to a laser system and more particularly to a laser system using ultra short laser pulses and a pulse shaper.
Commercially practical femtosecond lasers have been unavailable until recently. For example, lasers which can generate 10 femtosecond or less laser pulse durations have traditionally been extremely expensive, required unrealistically high electrical energy consumption (for extensive cooling, by way of example) and depended on laser dyes that had to be replenished every month thereby leading to commercial impracticality. The efficiency of sub-10 femtosecond lasers was not practical until the year 2000 because of the prior need for dyes and flash lamps instead of YAG and Ti: Sapphire crystals pumped by light or laser emitting diodes.
Ultrashort pulses are prone to suffer phase distortions as they propagate through or reflect from optics because of their broad bandwidth. There have been recent experimental attempts to shape the phase of ultrashort pulses since shaped pulses have been shown to increase the yield of certain chemical reactions and multiphoton excitation. Pulse shaping methods allowing synthesis of complex femtosecond optical waveforms are known. As usually practiced, the output waveform is determined by the Fourier transform (FT) of a spatial pattern transferred by a mask or a modulator array onto the dispersed optical spectrum. As an example, the FT optical pulse shaper is widely used for synthesis of complex femtosecond waveforms. In this geometry, the temporal profile of the output waveform is given by the FT of the mask pattern that is transferred onto the optical frequency spectrum of the pulse. FT pulse shaping was first demonstrated for use in simple pulses of tens of picoseconds in duration. Pulse shaping was then extended to the sub-100 femtosecond (fs) (10−15 second) time scale and demonstrated highly structured waveforms using microlithographically patterned pulse shaping masks. The introduction of liquid crystal modulator arrays and acousto-optic (A/O) modulators into FT pulse shapers led to computer programmable pulse shaping, with millisecond and microsecond reprogramming times, respectively, and widespread adoption of this technique.
These shaped pulses require a very large data set and in many cases, complex learning calculations for determining the pulse shaping characteristics for a particular application. The optimal pulse for the particular application is not known in advance. Since the variation shape of the possible pulse shapes is huge, scanning the entire parameter space is impossible and as such the optimized pulse shape could not have been predicted by theory. For a pulse shaper with N pixels, one can generate (P*A)N shaped pulses, where P and A are the number of different phases and amplitudes a pixel can take. If it is assumed 100 pixels, each taking 10 different amplitude values and 100 different phase values, the number of different pulses is of order of magnitude 10300. This dataset is extremely large, therefore, while in principle, the field exists to achieve the desired photonic transformation or excitation, finding it is a great challenge. It would be desirable for a system to control ultrashort pulses with a smaller dataset and operable to generate very complex pulse shapes that are optimal for the particular application.